Why is a direct comparative study best?

Clinical studies can either examine only one group of patients or compare this group with a second group of patients who are treated differently. A distinction is therefore made between comparative and non-comparative studies. The term "controlled studies" is also often used, because the terms "comparison group" and "control group" mean the same thing.

If only one group of patients (and therefore only one treatment) was examined in a study, it is only possible to determine whether the patients felt better after the treatment than before. However, such a before-and-after comparison is hardly meaningful. Imagine a study in which 100 people with a cold are given a sugar tablet. After one month, it is very likely that almost all of them will be "cured", but this corresponds to the natural course of the disease and does not prove the effectiveness of the treatment.

What is needed, therefore, is a comparison between two groups, one receiving the new treatment and one receiving the standard treatment. If both groups are treated at the same time and in the same place, this is called a direct, parallel comparison. Non-parallel comparison groups, on the other hand, are much less reliable because different conditions can create apparent differences. For example, if you examine Treatment A for colds in 2025 and compare it with the results of Treatment B in 2023, the informative value of this comparison is low. This is because the cold viruses, the people affected and the standard treatment may differ between the years, thus distorting the comparison between A and B.

A fair comparison of two drugs or treatment methods therefore requires direct comparative studies with a parallel design. This is best achieved through randomized controlled trials, as randomization also prevents the group comparison from being distorted by different patient characteristics.